ncomp_opls: Find the optimal number of OPLS components

In ricoderks/Rcpm: General functions for the Center of Proteomics and Metabolomics (LUMC)

Find the optimal number of OPLS components

Description

Find the optimal number of OPLS components based on the highest Q2 value.

Usage

ncomp_opls(Xi, Yi, prep, nrcv)

Arguments

Xi	X matrix N samples x K variables
Yi	Y matrix N samples x I variables (currently works only with single Y)
prep	preprocessing ("uv" for UV scaling, "mc" for mean centering, "no" for no preprocessing)
nrcv	number of folds in cross-validation

Details

Find the optimal number of OPLS components based on the highest Q2 value.

Value

Output is a list of three elements:

A	number of correlated components
ncox	number of orthogonal components in X
ncoy	number of orthogonal components in Y

Author(s)

E. Nevedmoskaya

Rico Derks

ricoderks/Rcpm documentation built on May 18, 2022, 7:49 a.m.